For Those Who Master Their Own Work

[zacnelson]

[RockChild56]

I always mix through the master bus compressors (mind you I don't do anything too severe on the master bus) - and I regularly check that I'm not over-compressing and I occasionally bypass them for a different perspective. The impression I get from my research is that most mix engineers mix through a master bus chain, even if they end up sending it elsewhere for mastering (in which case they will bypass some or all of those master bus effects)

Whatever you are doing Zac, I am going to do from now on.
Now if you could just tell us how you write your songs.

[zacnelson]

hehehee you are very flattering!

I'm happy to give you any tips if you want to ask, maybe about other plugins or sounds I use. Songwriting may be a harder thing to advise on, but I'd be happy to try

[andygabrys]

like what Zac said - and Gavin (the Element) - I think the sound of your master is primarily in your mix. So I compress and eq liberally on individual tracks, including high and low pass filters to reduce the amount of trash and look for that to be the sound of the mix.

The "mastering" I do is a) sweetening and b) perceived volume primarily.

So like Zac - the most standard chain I use is compressor, eq, limiter. My current favs are the UAD SSL G Buss comp, the UAD Brainworx BX Digital V2 Mid-side eq and usually the Slate FGX limiter.

On the comp - I time the attack and release to the track usually (again hitting it lightly around 1-3 db gain reduction unless I like to hear it working more on a more aggressive track). On something with more of a hip-hop beat, I usually go with a slower attack to get a lot of snap out of the drums.

on the eq - I often roll off bottom at 21 hz - I can't hear that low on my monitors necessarily but there is almost no system in the world that will reproduce that low so anything below that might as well be gone. And yes I have noticed you get a little more loudness out of tracks if you take care to get rid of the junk - if that's important. Since the eq is mid-side I often go for a little more stereo width and I usually take off a db or two with a notch or two between 100 hz and 1k - its those freqs esp around 400-500hz that ring and sound really boinky. I just take a little out to taste, varying freq from track to track. I just find I can get it sounding a bit sweeter to me. if I can't make it work on the master I return to the mix and do some work in the same areas on individual tracks, that often makes a huge difference.

then limit - the FGX is handy because there is a meter on it. I have a bunch of limiters, and once in a while I roll them all out and check out the different sounds. Some of them are easy to use, some of them are more finicky (like the Waves L316) but there are lots of good ones out there.

But beyond that, depending on the track I often use the final Mix Mix Buss plugin on tracks that have to be really loud and the UAD Studer on mixes that sound a little squeaky and don't seem to glue together. Which is usually stuff I mix for other people, as when I do my own tracks I arrange them in a way that works to my ear.

when I open up some of my older projects, I might have like 6 or 8 plugins on the master. With the way I listen now, I A/B all those plugins and usually end up stripping most of the them off. As I said for me - the sound is in the mix.

[zacnelson]

Lots of great advice there from Andy. There's one thing I would like to draw attention to in Andy's post; he said he only makes small adjustments to the EQ on the master bus. I think that is crucial. Many years ago when I was a worse mixer (and I'm not even very good now), I used to make very large EQ adjustments on the master bus, because I thought it was an efficient and logical way to get rid of ugly mids. But now I have realised that each part of the EQ spectrum has it's place, and it's not necessarily a problem if your mids are loud, the secret is to have the RIGHT mids! So I make sure I `share the frequencies around' between the instruments, for example I may cut the snare at 800hz, but NOT cut the vocals at 800hz, I will cut the strings or piano at 250hz but gently boost the acoustic guitar at 250hz, I will very aggressively cut the top end off electric guitar amps (especially with distorted sounds), but I will BOOST the top end a lot on lead vocals and cymbals. These specific EQ examples are not universal, there are times where you will approach it differently based on different genres. For example if you want a 60s or 70s drum sound you may boost the cymbals at 2khz or 3khz, and you would NEVER do that on country or modern rock. But the point I'm trying to make is that you can't resolve EQ balance by purely attacking the master bus; the secret to getting clarity and separation in the mix is to allow each instrument to `own' it's own territory in the spectrum. So much so that often if you were to solo certain individual instruments, they will sound very incomplete on their own. So if there is a section in a song where the piano or acoustic guitar is playing solo, you would EQ that section differently, and once the drums and bass etc come in that solo instrument would be narrowed down with a high-pass filter, less reverb, etc etc.

I hope I haven't meandered too much with these comments!!

[andygabrys]

I think that's right on Zac. I feel much the same.

It's sort of like peeling an onion - learning to hear and what to do in a mix.

The first layer is ham fisted overuse of everything.

A number of layers down you learn how to hear those things you mentioned, and also have some sort of mental reference (or a track) that lets you know if you are there.

If it was easy and everyone could do it all straight off....

[sansharbour]

Hey Zac, Andy, Russell, The Element

Thanks for contributing your tips and experience so that we all may benefit from trying new things.
I was hoping when I originally posted this was to get a discussion of how everybody gets around to making their material sound as good as possible with the individual plugins that each of us have.

Starting with great raw material, song or composition (Important)
Then to record effectively from the source would be the first step. No excuses here.
Mixing with various treatments to individual tracks. Listening closely in many environments, making decisions.

Lastly applying the home mastering techniques that make our material shine.

Don

[zacnelson]

Okay this is going to be a long post, but hopefully a good reference for anybody interested.

There are a LOT of articles and forum threads all over the internet that discuss the alchemy of mixing and you have to be careful when you read people's advice. The first thing you need to do is weigh up the advice you are reading by listening to the music that individual has created; you will find people spouting off very confidently about various topics or telling you certain must-haves in terms of gear or technique, and then you listen to their music and it's really poorly mixed. You'll get a lot of that on the Gearslutz forum. And PLEASE I urge you to apply the same criteria to me! I most definitely do NOT rate myself as a mixer, I am constantly second-guessing myself and doubting all my mix decisions. I think I mix better than SOME people, and I'm far more comfortable with pop/rock mixing, however I am a relative beginner when it comes to mixing orchestral - and it's a VERY different skill set. If you check out my soundcloud page you'll see some pop songs and some orchestral stuff, and if you hear anything in particular that you'd like to emulate, I'm more than happy to discuss specifics.

Coming from a pop and rock background, the focus is always about taming transients because everything is recorded very close and dry, and mixed very dry too. Therefore there is a lot of emphasis on compressors and limiters and trying to get the loudest most in-your-face mix, so someone like me is very comfortable mixing vocalists and drums etc, but then I turn to an orchestral mix and I feel like I'm completely lost! From what I've observed, there is a much greater emphasis on reverb in the sample world, there are some real experts on this forum who are masters of manipulating the sound field to create a sense that the orchestra was all in the same room.

COMPRESSION:

I don't think it is necessary to know all the different types of compressors and their characteristics; don't get bogged down with tangents like that, particularly when a lot of that is obsolete with software. A lot of the fabled compressors that pop mix engineers will reference are discussed in the context of slamming pop vocals and snare drums etc, whereas with orchestral music you have far less transients to deal with. There is also a lot more of a tolerance for overall dynamics in an orchestral mix, it is not expected to be at around 0dB from start to finish; the climaxes and emotional swells in orchestral music are largely dependent on allowing a huge volume range. When I do orchestral mixes I don't use compressors on ANY of my sub-groups except for percussion and choir (and only if I have a choir doing chants or staccato). BUT there is definitely a purpose for gentle compression on the mix buss. My main approach to this is actually not to control volume, but to `gel' the instruments together. Recently I got the ProAudioDSP plugin `DSM V2' (DSM means Dynamic Spectrum Mapper). It's a Plugin Alliance product, and I was lucky enough to get it for 70% off in their annual pre-Xmas sale. It has taken me ages to really understand, and it is capable of doing real damage when used incorrectly. It is a multiband compressor with dozens of bands and because of this it really has the unique ability to saturate everything, it's like when you make a sandwich and it's 6 inches tall and you press down and make it half that height, and all the ingredients blend together and intertwine, instead of sitting near each other. I don't want you to think I'm proposing a `squashed' mix in the rock mixing sense, but when you work with midi there are a lot of clunky volume and mod wheel moves which tend to sound a bit jumpy and `stepped' and the fast-acting multiple bands of compression really smooth out a lot of that, and it brings out the air and detail in the instruments without needing to add high eq.

With pop or rock music, I apply sometimes 3 or 4 compressors in a chain on the lead vocals, so that you can achieve a ridiculous amount of compression without it sounding unnatural. By feeding a compressed vocal into another compressor you can get better results than trying to do an enormous amount of compression with just one compressor. The best advice with lead vocals is to have fast attack and fast release settings on the compressor. With snare drums, fast attack and release will help to bring up the ping and after-hit resonance of the snare, whereas a slower attack and release will give you that short sharp snap (which can sound good with lots of reverb providing the decay of the snare, but is also a more dated sound).

LIMITING:

Please see my earlier post about my master bus limiting chain.

PANNING

Panning is a big issue in this day and age where we all depend on sample libraries for many of our sounds. Panning is more complicated with stereo virtual instruments and samples (eg strings, synth pads, choirs etc); many libraries have each patch as a big wide lush stereo spread, and when you are layering dozens of these together everything can become un-defined and messy. And even when some sample libraries have each section recorded `in place' they tend to still be only panned a little and the problems remain. It can be difficult to narrow the sounds because they sound so awesome on their own! But if you make each individual patch narrower, the cumulative effect of various instruments panned around the spectrum can sound FAR wider than when you stack lots of wide sounds on top of each other.

Originally I used to leave all the instruments really wide, but unfortunately they would each sound BRILLIANT individually but the final mix actually didn't sound very wide at all! So I discovered that if I make each instrument much more narrow, (even though it might sound a little too `mono' on it's own), the final effect with all those mono-ish instruments was SO MUCH WIDER when they're all playing together!!! It's hard to appreciate of course until the track is completed, when you start working on the track it can be a bit depressing, because there is a natural urge to want to hear the current solo-ed instrument as wide and full as it can be!

Also, instead of just panning stereo tracks to one side, I often use a plugin on the tracks called S1 Imager (by WAVES) which somehow makes panning more interesting, I have no idea what it does, I just play around with the sliders and make a panned instrument still have a sense of subtle width or depth or 3-D. Sometimes it sounds crap and I remove it, other times it sounds better. I just continually use `bypass' to compare the sound with S1 versus the sound of normal panning. I'm sorry I can't explain it better But most people seem to have one of the Waves plugin bundles and S1 appears in most of the bundles.

REVERB:

Reverb can be too obvious and very dated-sounding if it is too sparkly, you can get away with using more of it provided you roll off the high frequencies of the reverb. Also I put more reverb on things that are meant to be in the background.

VOCAL SEPARATION:

This is a massively important factor. An important thing to learn is how to separate your lead vocal from the backing vocals, vocal tags, and harmonies. There are a variety of things I do on all my songs that combine to achieve this. If you listen to a lot of my songs, there are a LOT of vocal tracks going on, interweaving and overlapping, often with the same voice doing different parts. Here are the secrets to making sure the lead vocal always shines in front of the others, despite the others being clearly audible.

1) If possible, use different singers. OR if it's the same singer, make sure a different part of the voice is used, ideally a blander less gritty tone with less dynamics.
2) Do HEAPS of takes of the BVs and harmonies and create a wall of vocals, which tends to make them more of an even blend, a smoother sound like the different between a solo violin and a full string section of 16 violins.
3) Put a bit more reverb on the non-lead vocals to push them further back in the mix.
4) Pan the non-lead vocals as a wide spread (100% left and 100% right) or if there is just a solo vocal doing a harmony or vocal tag, pan it slightly away from centre, or even use a doubler plugin.
5) Most importantly, take away the top-end sparkle and deep warmth; in particular removing the brightness of the non-lead vocals will trick the ear into thinking they are in the background, not the foreground.

EQ:

I covered a lot on this topic in a previous post, however here is some more explanation.

If you listen to good mixes there will be some things that are lower (ie deeper) and other things that are brighter; but not EVERY instrument is fully bright and fully deep.

The challenge is to choose which parts of your mix are going to occupy different territories frequency-wise. Often I will actually use a low-pass filter or a hi-shelf eq to REMOVE top end from some instruments, which ends up having the effect of actually brightening the overall mix because you can allow certain instruments to shine in the top end (especially vocals/choirs, cymbals, maybe brass--- give all those instruments lots of nice bright top end). Also you really need to high-pass heaps of instruments to allow the deep instruments to filll in the bottom end for you; the other instruments may sound a bit weird on their own but once the bass parts are there it all makes sense. If you don't know what a high-pass EQ is, it's just a standard technique where you remove or lessen the low end frequencies below a certain point. For example, with some lead instruments in a dense mix you could comfortably set a high-pass to around 200hz, which means the volume at frequencies below 200hz is dramatically reduced. This leaves more room for the bass guitar etc to occupy that area, I often boost a bass guitar at around 50hz.

Another important tip is to give different instruments their own zone in the mid range; you want everything to sound separate, but not to sound SMALL - the sum of the parts should make you think that EVERY part of the mix is as big as it could be, when in fact there are EQ holes that are dug out of one instrument to allow another instrument to focus on that sweet spot. Not every instrument sounds good in all parts of the spectrum; for example sometimes I will actually boost the low mids of a violin, whereas with a lot of percussion I will reduce the EQ in the low mids and bottom end. (And yet the percussion will still be perceived as `under' the mix, or deeper than the other melodic instruments, despite me cutting the bottom end eq on the percussion!). In an orchestral mix, sometimes to make a deep instrument like a cello or contrabass be heard, it works better to increase the upper mids and highs, instead of just increasing the volume of that instrument. Increasing the volume of the cello might just add a lot of boomy lows and low mids, whereas enhancing some of the bow noise and sparkle can lead the ears to the cello notes, and the deepness will be felt at the same time, but without being overpowering.

Lastly, please don't get diverted by any nonsense about mixing in-the-box or with outboard gear. That entire discussion is completely irrelevant and the answer to your quandary is NOT to go buying expensive outboard gear, or to blame your mix deficiencies on the fact you don't have some mysterious magical tool that is creating some huge gap in sound quality. Invest your money in good quality sound libraries and instruments, not in more mixing gear. A good piano library will make a hundredfold more improvement to your `sound' than some expensive compressor plugin which apparently does things so much better than a stock compressor that comes with your DAW.

There is absolutely no reason why somebody with enough practice and critical listening couldn't produce brilliant mixes with standard computer software, and of course with a selection of top sample libraries. Outboard gear in most people's hands would actually worsen the experience, because you can't be as nimble in shifting between projects and it takes up more room etc.

Also don't get side-tracked by the endless internet discussions about which converter sounds better or the supposed advantages of working at higher sample rates. None of that is going to make for a better mix.

[TheElement]

That is a long post Zac. lol Lot of good info in there.

Hey Don for me less is more. I keep tracks to minimum and try only to have a couple or so sounds going on at the same time. I also have great synths..Massive, Nexus2, A.N.A. , Sylenth1 and for kicks..the KICK. Also got lots of decent instrument samples from Sample Magic.

My DAW of choice is Cubase. I use Cubase plugins but also got a few other ones like Sausage Fattener and The Glue compressor.

I keep enough headroom and eq everything to try and fit it all together so song sounds as clear as I can get it.

After all that is done I export a mix down with enough headroom (more than -6 db) to a wav.

Then its time for mastering. I use the online mastering service LANDR. I pay for unlimited wav files per month. I take the mastered wav file, load it into Cubase and make a 320 kbps mp3 for submitting to listings.

Thats all to it. Thats my system.

[sansharbour]

Thanks everyone for a world of information.

Don